You might want to look at SecuritySage for some configuration details.
http://www.securitysage.com/guides/postfix_uce.html
I just got some mail bounced by an ISP using this setup and after reviewing the details, it looks like the system is set up backwards: It does its own all-or-nothing header checks that SA already takes care of, causing heavy-handed rejections of false positives, instead of letting SA look the message over and using the combined score of several indicators to make the decision.
In my case a message to a cogeco.ca customer got rejected for the regex "^Subject: .*Get Ready For". The subject line in question was "How to get ready for a match". (This was sent to all my online teammates.) Here's the Postfix header checks rules used:
<http://www.securitysage.com/files/header_checks>
(On my own servers I run SA from sendmail via MIMEDefang and reject spam with a score of 10 or higher, and add the spam score to the subject line for stuff between 5 and 10 for MUA filtering. The only SA hit I got on the message in question was NO_REAL_NAME, and my AWL kicked the thing down to a -4.9 score.)
Poking around on Cogeco's website, I can find no indication that they offer spam filtering, and certainly nothing to suggest that they delete mail bound for their customers with no customer control. One would think this would be a selling feature of their service and that they'd advertise its existence.
<http://cogeco.ca/en/internet_support_o.html>
------------------------------------------------------- This SF.Net email sponsored by: ApacheCon 2003, 16-19 November in Las Vegas. Learn firsthand the latest developments in Apache, PHP, Perl, XML, Java, MySQL, WebDAV, and more! http://www.apachecon.com/ _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
